The following content has been provided by the University of Erlangen-Nürnberg.
So, welcome to the lecture Interventional Medical Image Processing.
Unfortunately I wasn't able to teach last week, so we had some kind of refresher courses
on linear algebra and on homogeneous coordinates, two concepts that we are going to use from
now on without any comment.
Okay.
you this should have been just a refresher course last week. Today we want to start with
the topics we want to consider. And the first question is, what is intervention? Who is
the lecturer actually? My name is Joachim Horneger for those of you who don't know
me. I'm in charge of the pattern recognition lab and my major application field is medical
image processing as you might know. And I'm guiding you through the summer semester lecture
and I hope that I can teach most of the lectures for you. They will be recorded. You can also
have access to previous lectures or lectures of previous years. And as I said, we want
to adjust this semester a little bit the topics to make the lecture more up to date regarding
current developments in the field. The question now is why is the lecture called interventional
medical image processing? In the winter semester I teach diagnostic medical image processing
and the summer interventional. So what's the difference? Diagnostic medical image processing
basically means there is a highly paid technician. Welcome. You can sit here if you want.
Don't smile, it's recorded. There is a technician and the technician puts the patient into the
system. The patient gets an x-ray image, the patient gets a CT image, the patient gets
an MR image. Then the technician sits there and adjusts the image quality and works on
the pictures. An hour later or the next day or whenever a radiologist will sit down and
do a diagnosis on the basis of these images. In interventional that was the diagnostic
part. We talked about image pre-processing in winter. We talked about three-dimensional
reconstruction. We talked about combining different images in a joint coordinate system.
The summer semester we will look into a completely different application domain. The imaging
system, the imaging device will be used by the doctor himself. So the doctor will use
the system while he is treating the patient. So he's doing surgery and using the medical
imaging equipment during surgery. Or he is doing an intervention in the sense that he's
fixing up something in the cerebral vessel system in the brain without surgery but by
just going through the vessels and using x-ray to see where he is and what he actually does.
You can imagine this type of procedure requires a completely different system. I mean if a
CT system breaks and you have to reboot the system in a diagnostic phase, you lose some
time but it's not life-threatening for the patient. You have to tell him, look we have
to do it again. In an intervention if you know the catheter is right in the coronary
arteries of the heart and all over a sudden the x-ray system does not deliver any pictures
anymore, you are there with your catheter and you do not see anymore what you are actually
doing. So reliability, safety is an issue. Or speed, computational power, the computational
speed is an issue. Also the user interface is an issue. I mean you don't have a highly
qualified technician but you have a poorly technically equipped doctor running the system
and he wants to have one button, right? It's just confidential information. I'm doing this
type of jokes all the time. You will get used to it. So the system has to be very easy to
access because the doctor does not want to concentrate on the interface to the x-ray
machine but he wants to concentrate on the patient and how the patient can be properly
treated.
So there are a few demands which are completely different in the interventional setup like
in the diagnostic setup. Good. A few questions up front. Can we shift the time for the lecture?
The answer is no. Yeah? So maybe you have discussed this last, did you see this slide
last week already? Okay so I can skip that. Anyways. Good. And then let me just check
whether my hard disk is already up here. It's still not showing my disk. That's weird. But
Presenters
Zugänglich über
Offener Zugang
Dauer
01:28:03 Min
Aufnahmedatum
2012-04-23
Hochgeladen am
2012-04-26 17:05:38
Sprache
en-US
This lecture focuses on recent developments in image processing driven by medical applications. All algorithms are motivated by practical problems. The mathematical tools required to solve the considered image processing tasks will be introduced.